优化YOLOv12n实现红外小目标实时检测，显著提升精度与速度。
TY-RIST: Tactical YOLO Tricks for Real-time Infrared Small Target Detection
- 采用步长感知骨干网络和高分辨率检测头，增强小目标特征提取。
- 引入级联坐标注意力与剪枝策略，精度提升同时推理快至123帧/秒。
- 适合军事侦察、安防监控等需快速精准识别小目标的场景。
红外小目标检测在国防与监控中至关重要，但面临目标特征微弱、杂波环境误报、低显著性漏检及计算成本高等挑战。为此，我们提出TY-RIST，一种基于YOLOv12n优化的架构，包含：(1) 步长感知骨干网络与细粒度感受野，(2) 高分辨率检测头，(3) 级联坐标注意力模块，(4) 分支剪枝策略，使计算量降低约25.5%，精度小幅提升，支持单卡123 FPS实时推理。同时引入归一化高斯沃瑟斯坦距离（NWD）提升回归稳定性。在四个基准数据集上，20种模型对比实验显示，[email protected] IoU提升7.9%，精确率+3%，召回率+10.2%。第五个跨数据集验证进一步证明其强泛化能力。更多资源见https://www.github.com/moured/TY-RIST。
原文摘要 · Abstract (English)
Infrared small target detection (IRSTD) is critical for defense and surveillance but remains challenging due to (1) target loss from minimal features, (2) false alarms in cluttered environments, (3) missed detections from low saliency, and (4) high computational costs. To address these issues, we propose TY-RIST, an optimized YOLOv12n architecture that integrates (1) a stride-aware backbone with fine-grained receptive fields, (2) a high-resolution detection head, (3) cascaded coordinate attention blocks, and (4) a branch pruning strategy that reduces computational cost by about 25.5% while marginally improving accuracy and enabling real-time inference. We also incorporate the Normalized Gaussian Wasserstein Distance (NWD) to enhance regression stability. Extensive experiments on four benchmarks and across 20 different models demonstrate state-of-the-art performance, improving mAP at 0.5 IoU by +7.9%, Precision by +3%, and Recall by +10.2%, while achieving up to 123 FPS on a single GPU. Cross-dataset validation on a fifth dataset further confirms strong generalization capability. Additional results and resources are available at https://www.github.com/moured/TY-RIST
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。